Carotid endarterectomy to correct asymptomatic carotid stenosis: Ten years

Transcription

1 Carotid endarterectomy to correct asymptomatic carotid stenosis: Ten years later David Rosenthal, M.D., Randal Rudderman, M.D., Edgar Borrero, M.D., David H. Hafner, M.D., Garland D. Perdue, M.D., Pano A. Lamis, M.D., Michael D. Clark, M.D., and Wayne W. Daniel, Ph.D., Atlanta, Ga. The ideal management of the patient with an asymptomatic stenosis of the extracranial internal carotid artery remains controversial. The purpose of this article was to evaluate the effects of prophylactic carotid endarterectomy (CE) done to treat asymptomatic carotid stenosis (greater than 50% diameter reduction by angiography) 10 years later. In 1976, 42 prophylactic CEs were performed. There were no postoperative deaths or strokes. During 10-year follow-up two strokes occurred in the operated hemisphere; one stroke was fatal and was due to an intracranial hemorrhage, whereas the other stroke was thromboembolic in origin. Two other patients suffered strokes in the contralateral hemisphere and seven patients had transient ischemic attacks in the contralateral hemisphere, which necessitated CE. The survival rate at the end of the study period by life-table analysis was 57% (mean 8.7 years). Sixteen late deaths occurred, with coronary arteryrelated disease the most common cause of death. This review with actual 10-year followup demonstrated that prophylactic CE may be performed with minimal risk, that late stroke in the operated hemisphere was negligible, and that long-term survival was similar to that of a comparable age-matched population, possibly because late deaths attributed to stroke were reduced. On the basis of long-term follow-up, CE to treat asymptomatic high-grade carotid stenoses appears to be indicated in appropriate patients. (J VAse SuRG 1987;6: ) The ideal therapy for the patient with an asymptomatic stenosis of the extracranial internal carotid artery remains controversial. Several alternatives of treatment have been advocated, including prophylactic carotid endarterectomy (CE), interval noninvasive testing, and "waiting" for the occurrence of neurologic symptoms. This controversy will probably continue until information is obtained that can (1) clarify the natural history of untreated asymptomatic carotid stenosis, (2) evaluate factors that may affect therapeutic results, and (3) assess the long-term consequences of prophylactic CE. The purpose of this article was to evaluate the effects of prophylactic CE for asymptomatic carotid stenosis 10 years after operation. From the Departments of Vascular Surgery, Georgia Baptist Medical Center (Drs. Rosenthal, Rudderman, Borrero, Lamis, Clark, and Daniel), and Emory University Hospital (Drs. Hafner and Perdue). Presented at the Eleventh Annual Meeting of the Southern Association for Vascular Surgery, Scottsdale, Ariz., Jan , Reprint requests: David Rosenthal, M.D., 315 Boulevard NE, Suite 412, Atlanta, GA MATERIAL AND METHODS In 1976, 42 patients at Georgia Baptist Medical Center and Emory University Hospital, Atlanta, Georgia, underwent prophylactic CE to treat asymptomatic stenosis (greater than 50% diameter reduction as documented by arteriography) of the internal carotid artery. Patients were defined as asymptomatic if the cerebral hemisphere supplied by the operated artery was free of transient ischemic attacks (TIAs), monocular visual symptoms, or prior infarction. The group consisted of 31 patients who were totally asymptomatic, as well as 11 patients with symptoms who had undergone contralateral CE. Of the 42 patients, 29 were men. The mean age was 61 years (range 49 to 77 years). Twenty-one patients had a history consistent with coronary artery disease. Hypertension was present in 18 patients, whereas seven patients had diabetes mellitus, and 23 patients had a history of cigarette smoking. Twenty-nine patients (69%) had a bruit audible over the asymptomatic carotid bifurcation. The presence of this bruit initiated further diagnostic studies, which included Kartchncr oculoplethysmography (OPG) and carotid phonoangiography (CPA) 226

2 Volume 6 Number 3 September 1987 Prophylactic carotid endarterectomy: Ten years later 227 Percent Survival 100! 90! 80' 70' 60, 50' 40' 30' 20' 10' 0 (Mean Survival = years),, Fig. 1. Cumulative late survival. performed on 18 of the 29 patients. "Hemodynamically significant" stenoses by OPG/CPA were demonstrated in 15 of the patients, whereas another 11 patients underwent arteriography without OPG/CPA studies. The remaining 13 patients had their stenotic arteries identified as a chance finding at carotid arteriography performed because of contralateral hemisphere symptoms (10 patients) or during aortic/coronary arteriography before a major cardiovascular reconstruction (three patients). Arteriographically proven stenoses between 50% and 75% of the arterial diameter were present in 38% of patients (16) undergoing prophylactic CE, whereas 62% of patients (26) had stenoses between 75% and 99%. Seven percent of patients (three) had total occlusion of the contralateral carotid artery, 19% (eight patients) had stenoses greater than 50% diameter reduction, and 12% (five patients) had erosions of the contrallateral carotid artery. Duplex scanning and spectral analysis were used to evaluate carotid artery patency in 18 patients 10 years after CE. Follow-up data were obtained from office and hospital records, patients, their surviving family members, or physicians. For patients who had died between operation and review, the patients' neurologic status, as well as cause of death, was documented by contacting family members, medical records, or death certificates whenever possible. Followup data concerning neurologic morbidity, as well as survival, were tabulated with actual years used as follow-up. Long-term follow-up ranged to 10 years (estimated mean survival time was ± 5.5 months, median greater than 111 months). Six patients were lost to long-term follow-up--one at 4 years, three at 5 years, and two at 9 years (Table I). # ' 90' 85' 80' 75 3O Cumulative Stroke-free Rate (%) --- Ipstlateral stroke Total strokes Fig. 2. Cumulative stroke-free rate :- RESULTS Forty-two patients underwent CE to treat asymptomatic, high-grade stenoses of the internal carotid artery. There were no deaths and no strokes in the immediate postoperative period (less than 30 days). Two patients had transient postoperative neurologic deficits within 6 hours of operation. One patient experienced monocular amaurosis fugax and the other a weak hand grip. Both deficits resolved without consequence within 3 hours. Fig. 1 (Table I) displays the cumulative individual late survival of the patients who had operation. The

3 228 Rosenthal et al. Journal of VASCULAR SURGERY Table I. Cumulative 10-year survival Lost to Interval since No. of patients Cumulative Cumulative survival follow-up operation (yr) entering interval deaths rate (%) (cumulative) Table II. Causes of late death (N = 16) Cause No. % Cardiac 9 56 Cancer 4 25 Stroke* 1 6 Unknown 2 13 *Intracranial hemorrhage. estimated 10-year survival rate of the group was 57%. There were 16 late deaths and the causes of death are summarized in Table II. The overall long-term stroke rate, including both ipsilateral and contralateral events, is delineated in Fig. 2 and Table III. Two strokes occurred in the operated hemisphere during 10-year follow-up; one fatal stroke was an intracranial hemorrhage documented by CT brain scan, whereas the other was thought to be thromboembolic in origin (Table IV). Two patients had TIAs in the operated hemisphere during the 10-year follow-up. Reoperation was necessary for recurrent stenosis in one, whereas the other patient was treated successfully with antiplatelet agents. Two patients suffered thromboembolic strokes in the contralateral hemisphere during follow-up (Table IV). At carotid arteriography, a high-grade stenosis in one and an excavated plaque in the other patient were identified as the causes of the strokes. Both patients regained nearly all neurologic function after these stroke events and underwent uneventful CE 6 weeks later. Seven other patients had TIAs in the contralateral hemisphere and also underwent uneventful CE. Thirteen patients in 1976 had contralateral carotid disease treatable by surgical therapy identified at arteriography: eight stenoses greater than 50% diameter reduction and five carotid bifurcation ero- sions. Twelve of the 13 patients (92%) ultimately required contralateral CE during the 10-year followup: the indications for operation included stroke without antecedent TIAs (two patients), hemispheric TIAs (seven patients), and progression ofasymptomatic high-grade stenosis (three patients). Eighteen patients were available for duplex scanning and spectral analysis 10 years after prophylactic CE. Sixteen of them had minimal evidence of recurrent carotid stenosis (less than 25% diameter reduction), whereas changes indicating greater than 50% diameter stenosis were observed in the remaining two. One patient with asymptomatic late recurrent carotid stenosis underwent arteriography, which demonstrated an approximate 50% smooth stenosis of the intemal carotid artery. The patient was given antiplatelet agents with duplex scanning scheduled at 6-month intervals. The other patient refused further diagnostic studies. DISCUSSION The value of CE in the management of the patient with an asymptomatic carotid stenosis remains controversial and poorly documented. Surprisingly little information is available on the natural history of patients with asymptomatic carotid stenoses because up until recently only contrast angiography could be used to document accurately the evolution of carotid bifurcation disease. With the advent of Doppler ultrasound techniques and digital arteriography, interesting data on the natural history of the untreated asymptomatic carotid stenosis have begun to accumulate. Several recent reports have reproduced the same findings, that is, that untreated high-grade stenosis of the carotid artery causes stroke. Hertzer et al.1 evaluated 195 patients who underwent intravenous

4 Volume 6 Number 3 September 1987 Prophylactic carotid endarterectomy: Ten years later Percent Survival Normal population t PCE I I I I I I I ~ --~ Fig. 3. Cumulative late survival of age-matched patient populations. PCE = prophylactic carotid endarterectomy. Table III. Cumulative 10-year incidence of ipsilateral and contralateral stroke Interval since No. of patients No. of strokes Interval stroke Cumulative stroke Cumulative stroke-~ee Lost to operation (yr) entering interval in interval rates (%) rate (%) rate (%) follow-up digital subtraction arteriography and found that the risk of stroke was concentrated among patients with unilateral 70%, or bilateral 50%, carotid stenosis; the 5-year cumulative stroke rate among these nonoperated patients was approximately 25%/ With the use of B-mode ultrasonography and spectral analysis, Roederer et al.2 reported an approximate 50% incidence of neurologic events among patients with stenoses exceeding 80%, whereas Chambers and Norris 3 reported a 15% annual neurologic event rate in patients with carotid stenoses exceeding 75% compared with a 3% annual stroke rate in patients with lesser stenoses. Such observations are consistent with the concept that high-grade carotid lesions often undergo subintimal hemorrhage or plaque necrosis, which may precipitate either cerebral microemboli or sudden ocdusion of the internal carotid artery 4-6 A review of the 10-year survival rate of our patients compared with that of comparable age- and sex-matched populations without extracranial arterial occlusive disease 7 was performed. It was interesting that although the patients who had prophylactic CE had "significant" carotid artery disease, their longterm estimated survival approached that of a comparable patient population without arterial disease, possibly because of a lower incidence of stroke as a cause of late death (Fig. 3). The incidence of stroke 10 years after prophylactic CE according to life-table estimates was remarkably low (Fig. 2). Indeed, only one patient had a cerebral infarction on the side of prophylactic CE that was thromboembolic in origin, whereas another patient suffered a cerebral hemorrhage on the operated side. Two other patients had TIAs in the operated hemisphere, but "redo" CE was necessary in

5 230 Rosenthal et al. Journal of VASCULAR SURGERY Table IV. Neurologic status (N = 42) No. % Operated hemisphere Stroke (one infarct/one 2 5 hemorrhage) TIA 2 5 Nonoperated hemisphere Stroke (infarcts) 2 5 TIA 7 16 Asymptomatic TIA = transient ischemic attack. only one because antiplatelet drugs rendered the other patient asymptomatic. The development of neurologic symptoms in the contralateral hemisphere, after carotid artery disease was documented in 1976, was alarming. Two patients suffered completed strokes without antecedent TIAs, whereas seven other patients had TIAs and required CE. These patients seem to be at high risk for compromise of the central nervous system and should be observed closely with semiannual duplex scanning combined with spectral analysis and treated aggressively as indicated. The durability of prophylactic CE seems to be verified as 88% of patients evaluated by duplex scanning and spectral analysis 10 years after operation had patient arteries with minimal evidence of recurrent stenosis. If prophylactic CE can be accomplished with no perioperative deaths and minimal morbidity (less than 2% postoperative permanent neurologic deficit), it would seem to be a small price to pay for longterm stroke protection. Although the reported perioperative complication rates in major referral centers for CE are less than 4%, sl city-wide audits in Springfield, Illinois la and Cincinnati, Ohio ~2 have shown alarmingly high perioperative stroke and/or mortality rates of 21% and 11%, respectively. Because of these conflicting reports and information, controversy over the role of prophylactic CE to treat asymptomatic carotid stenosis continues. Hopefully, the ongoing randomized controlled Veterans Administration Cooperative Study will resolve the role of prophylactic CE and the polarizing issue of operation vs. no operation. However, this review, based on actual 10-year follow-up, demonstrated that prophylactic CE may be performed with minimal risk, that late stroke in the operated hemisphere was negligible, and that long-term survival was similar to that of a comparable population, possibly because late deaths attributed to stroke were reduced. CE performed to correct asymptomatic high-grade carotid stenoses appears to be indicated in appropriate patients. REFERENCES 1. Hertzer NR, Flanagan RA, Beven EG, O'Hara PJ. Surgical versus nonsurgical treatment of symptomatic carotid stenosis. Ann Surg 1986;204: Roederer GO, Langlois YE, Luisiani L, et al. Natural history of carotid artery disease on the side contralateral to endarterectomy. J VASC SURG 1984;1: Chambers BR, Norris JW. The case against surgery for asymptomatic carotid stenosis. Stroke 1984; 15: Lusby RJ, Ferrell LD, Ehrenfeld WK, et al. Carotid plaque hemorrhage: its role in production of cerebral ischemia. Arch Surg 1982;117: Imparato AM, Riles TS, Minter R, Baumann G. The importance of hemorrhage in the relationship between gross morphologic characteristics and cerebral symptoms in 376 carotid artery plaques. Ann Surg 1983;197: Lord RSA. Late survival after carotid endarterectomy for transient ischemic attacks. J VAsc SURG 1984;1: America in transition: An aging society. US Department of Commerce Bureau of the Census, series P-23, 1983, No Hertzer NR, AveUone JC, Farrell CJ, et al. The risk of vascular surgery in a metropolitan community. J VAsc SURG 1984; 1: Krupski WC, Effeney DJ, Goldstone J, et al. Carotid endarterectomy in a metropolitan community: comparison of results for three institutions. Surgery 1985;98: Slavish LG, Nicholas GF, Gee W. Review of a community hospital experience with carotid endarterectomy. Stroke 1984;15: Easton JD, Sherman DG. Stroke and mortality rate in carotid endarterectomy: 228 consecutive operations. Stroke 1977;8: Brott T, Thallinger K. Carotid endarterectomy in Cincinnati 1980: indications and morbidity in 431 cases. Neurology 1983;33:93-5.